Lazarus debugger fail in Ubuntu 13.04
lazarus / 12-Debugger

----------------------------------------------------------------------
miguel-laz · 2013-04-29 19:07

Just installed Lazarus 1.0.8 on Ubuntu 13.04 (64bit) and trying to
debug a simple program, the debugger stops at the breakpoint marked
but when you stop debugging (red square button) displays a debugger
error window indicating "The debugger entered into a failed state." In
windows7 32bit works perfectly.

This is the simple program:
Code: [Select]unit Unit1;

{$mode objfpc}{$H+}

interface

uses
  Classes, SysUtils, FileUtil, Forms, Controls, Graphics, Dialogs,
StdCtrls;

type

  { TForm1 }

  TForm1 = class(TForm)
    Edit1: TEdit;
    procedure Edit1Enter(Sender: TObject);
  private
    { private declarations }
  public
    { public declarations }
  end;

var
  Form1: TForm1;

implementation

{$R *.lfm}

{ TForm1 }

procedure TForm1.Edit1Enter(Sender: TObject);
var str:string;
begin
     str:='Un texto';
     TEdit(Sender).Text:=str;
end;

end.

The breaking point is on the line that says "str:='Un texto';":

The version of gdb is GNU gdb (GDB) 7.5.91.20130417-cvs-ubuntu

Has anyone tried Lazarus on Ubuntu 13.04 (64bit)?
Can anyone reproduce the error?

Thanks in advance

----------------------------------------------------------------------
Martin_fr · 2013-04-29 19:22

Can you try to produce a log and attach it:
http://wiki.lazarus.freepascal.org/GDB_Debugger_Tips#Log_info_for_debu
g_session

----------------------------------------------------------------------
marcov · 2013-04-29 21:27

Btw, gdb 7.6 came out in recent days. Listed some MII improvements.

----------------------------------------------------------------------
miguel-laz · 2013-04-30 16:00

As the text exceeds 20000 characters the whole message is in the
attachment:

OK, The following text corresponds to the output of the debugger until
the breakpoint corresponding to the line "str:='Un texto';":

Code: [Select]=thread-group-added,id="i1"
(gdb)
<-gdb-set confirm off>
^done
(gdb)
<-gdb-set new-console off>
^error,msg="No hay tabla de s\303\255mbolos cargada. Use la orden
\302\253file\302\273."
(gdb)
<set width 50000>
&"set width 50000\n"
=cmd-param-changed,param="width",value="50000"
^done
(gdb)
<-gdb-version>
~"GNU gdb (GDB) 7.5.91.20130417-cvs-ubuntu\n"
~"Copyright (C) 2013 Free Software Foundation, Inc.\n"
~"License GPLv3+: GNU GPL version 3 or later
<http://gnu.org/licenses/gpl.html>\nThis is free software: you are
free to change and redistribute it.\nThere is NO WARRANTY, to the
extent permitted by law.  Type \"show copying\"\nand \"show warranty\"
for details.\n"
~"This GDB was configured as \"x86_64-linux-gnu\".\nPara las
instrucciones de informe de errores, vea:\n"
~"<http://www.gnu.org/software/gdb/bugs/>.\n"
^done
(gdb)
<-gdb-set env COMPIZ_BIN_PATH=/usr/bin/>
^done
"
"      "Text in file"
"
(gdb)
<-stack-list-frames 0 4>
^done,stack=[frame={level="0",addr="0x0000000000485316",func="EDIT1ENT
ER",file="unit1.pas",fullname="/home/miguel/Programacion/LazarusTest1/
unit1.pas",line="35"},frame={level="1",addr="0x000000000061b77d",func=
"CONTROLS_TWINCONTROL_$__DOENTER"},frame={level="2",addr="0x00007ffff7
fbfbd0",func="??"},frame={level="3",addr="0x000000000064f1c3",func="ST
DCTRLS_TCUSTOMEDIT_$__DOENTER"},frame={level="4",addr="0x0000000000620
dc3",func="CONTROLS_TWINCONTROL_$__CMENTER$TLMESSAGE"}]
(gdb)
<-stack-list-arguments 1 0 0>
^done,stack-args=[frame={level="0",args=[{name="this",value="0x7ffff7f
be5d0"},{name="SENDER",value="0x7ffff7fbfbd0"}]}]
(gdb)
<-stack-list-locals 1>
^done,locals=[{name="STR",value="0x0"}]
(gdb)


The following text corresponds to the output of the debugger when
debugging stop (red square button) and show the window debugger error.

Code: [Select]<kill>
&"kill\n"
=thread-exited,id="3",group-id="i1"
=thread-exited,id="2",group-id="i1"
=thread-exited,id="1",group-id="i1"
=thread-group-exited,id="i1"
^done
(gdb)
<info program>
&"info program\n"
~"El programa que ha sido depurado no estaba corriendo.\n"
^done
(gdb)

----------------------------------------------------------------------
miguel-laz · 2013-04-30 16:07

The first error message says something like:
^error,msg="No hay tabla de s\303\255mbolos cargada. Use la orden
\302\253file\302\273."

which in English means something like:
^error,msg="No symbol table loaded. Use the command?. ".

----------------------------------------------------------------------
Martin_fr · 2013-04-30 17:17

Yes, but IIRC that error msg can occur.

I will need the entire log file.

Also, I see that you use a translated gdb (not English). I have see
similar cases, and some messages that the IDE looks for in English,
are also translated. Then the IDE can not work with that gdb.

----------------------------------------------------------------------
macmike · 2013-04-30 17:55

I've been using Lazarus since v9.3 on Ubuntu without problems until I
upgraded to Ubuntu 13.04 (32-bit) on the weekend and now I get this
problem too

----------------------------------------------------------------------
Martin_fr · 2013-04-30 18:03

Again: log file please.

Lazarus is currently tested with gdb (english) from 6.3 to 7.5
I will hopefully be able to test tests for 7.6 soon But I don't know
how soon.

GDB 7.6 (even english) may have changes, in the response output of
some commands. Then the IDE does not understand it.

In addition. If gdb is translated, it may or may not work. I have seen
translated gdb working, because the IDE used fallbacks, if it got
replies that it could not read (and luckily the fallbacks had enough
keywords to be understood). But beeing already forced to fallbacks
makes it more likely to fail on other changes.

----------------------------------------------------------------------
macmike · 2013-04-30 18:09

I'm using the English versions. Where can I find the log file?
(sorry for the dumb question!)

----------------------------------------------------------------------
Martin_fr · 2013-04-30 18:23

Quote from: macmike on April 30, 2013, 08:09:24 pmI'm using the
English versions. Where can I find the log file?
(sorry for the dumb question!)

As I wrote earlier in this thread:
http://wiki.lazarus.freepascal.org/GDB_Debugger_Tips#Log_info_for_debu
g_session

----------------------------------------------------------------------
miguel-laz · 2013-04-30 19:19

OK, this is the log file
Code: [Select]TMainIDE.ParseCmdLineOptions:
  PrimaryConfigPath="/home/miguel/.lazarus"
  SecondaryConfigPath="/etc/lazarus"
TIDEAnchorDockMaster.MakeIDEWindowDockSite MainIDE:TMainIDEBar
TIDEAnchorDockMaster.LoadUserLayout anchordocklayout.xml
TIDEAnchorDockMaster.LoadUserLayout restoring ...
TMainIDE.DoLoadLFM Creating designer for hidden component of
/home/miguel/Programacion/LazarusTest1/unit1.pas
TMainIDE.DoOpenProjectFile select form in designer: Form1:TForm1
TDesigner
[TMainIDE.DoRunProject] A
TMainIDE.DoInitProjectRun A True 0
TMainIDE.DoInitProjectRun B
TLazPackageGraph.CheckIfCurPkgOutDirNeedsCompile  Compiler custom
params changed for FCL 1.0.1
  Old=" -MObjFPC -Scghi -O1 -g -gl fcllaz.pas"
  Now=" -MObjFPC -Scghi -O1 -g -gl -O2 -g- -Xs fcllaz.pas"
  State file="/usr/share/lazarus/1.0.8/packager/units/x86_64-
linux/FCL.compiled"
TLazPackageGraph.CheckIfPackageNeedsCompilation normal output dir is
not writable: /usr/share/lazarus/1.0.8/packager/units/x86_64-linux/
TLazPackageGraph.GetFallbackOutputDir  FCL:
/home/miguel/.lazarus/lib/units/x86_64-linux
TParsedCompilerOptions.SetOutputDirectoryOverride
New=/home/miguel/.lazarus/lib/units/x86_64-linux
TLazPackageGraph.CheckIfCurPkgOutDirNeedsCompile  Compiler custom
params changed for LazUtils 1.0
  Old=" -MObjFPC -Scghi -O1 -g -gl lazutils.pas"
  Now=" -MObjFPC -Scghi -O1 -g -gl -O2 -g- -Xs lazutils.pas"
  State file="/usr/share/lazarus/1.0.8/components/lazutils/lib/x86_64-
linux/LazUtils.compiled"
TLazPackageGraph.CheckIfPackageNeedsCompilation normal output dir is
not writable:
/usr/share/lazarus/1.0.8/components/lazutils/lib/x86_64-linux/
TLazPackageGraph.GetFallbackOutputDir  LazUtils:
/home/miguel/.lazarus/lib/LazUtils/lib/x86_64-linux
TParsedCompilerOptions.SetOutputDirectoryOverride
New=/home/miguel/.lazarus/lib/LazUtils/lib/x86_64-linux
TLazPackageGraph.CheckIfCurPkgOutDirNeedsCompile  Compiler custom
params changed for LCLBase 1.0.8
  Old=" -MObjFPC -Scghi -O1 -g -gl alllclunits.pp"
  Now=" -MObjFPC -Scghi -O1 -g -gl -O2 -g- -Xs alllclunits.pp"
  State file="/usr/share/lazarus/1.0.8/lcl/units/x86_64-
linux/LCLBase.compiled"
TLazPackageGraph.CheckIfPackageNeedsCompilation normal output dir is
not writable: /usr/share/lazarus/1.0.8/lcl/units/x86_64-linux/
TLazPackageGraph.GetFallbackOutputDir  LCLBase:
/home/miguel/.lazarus/lib/LCLBase/units/x86_64-linux
TParsedCompilerOptions.SetOutputDirectoryOverride
New=/home/miguel/.lazarus/lib/LCLBase/units/x86_64-linux
TLazPackageGraph.CheckIfCurPkgOutDirNeedsCompile  Compiler custom
params changed for LCL 1.0.8
  Old=" -MObjFPC -Scghi -O1 -g -gl lcl.pas"
  Now=" -MObjFPC -Scghi -O1 -g -gl -O2 -g- -Xs lcl.pas"
  State file="/usr/share/lazarus/1.0.8/lcl/units/x86_64-
linux/gtk2/LCL.compiled"
TLazPackageGraph.CheckIfPackageNeedsCompilation normal output dir is
not writable: /usr/share/lazarus/1.0.8/lcl/units/x86_64-linux/gtk2/
TLazPackageGraph.GetFallbackOutputDir  LCL:
/home/miguel/.lazarus/lib/units/x86_64-linux/gtk2
TParsedCompilerOptions.SetOutputDirectoryOverride
New=/home/miguel/.lazarus/lib/units/x86_64-linux/gtk2
TMainIDE.DoInitProjectRun
ProgramFilename=/home/miguel/Programacion/LazarusTest1/project1
[TCmdLineDebugger] Debug PID: 5117
TGDBMIDebugger.ProcessResult Error: ,msg="No hay tabla de
s\303\255mbolos cargada. Use la orden \302\253file\302\273."
[Debugger] Log output: &"set width 50000\n"
[WARNING] Debugger: Unexpected async-record: =cmd-param-
changed,param="width",value="50000"
[Debugger] Running GDB version: GDB
[TDebugger.SetFileName]
"/home/miguel/Programacion/LazarusTest1/project1"
[TMainIDE.DoRunProject] B TGDBMIDebugger
[TMainIDE.DoRunProject] END
TGDBMIDebugger.StartDebugging
WorkingDir="/home/miguel/Programacion/LazarusTest1/"
TGDBMIDebugger.ProcessResult Error: ,msg="El s\303\255mbolo
\302\253FPC_THREADVAR_RELOCATE_PROC\302\273 no est\303\241 en el
contexto actual."
[Debugger] Log output: &"info functions FPC_CPUINIT\n"
[WARNING] Debugger: Unexpected async-record: =cmd-param-
changed,param="args",value=""
[Debugger] Log output: &"set inferior-tty /dev/pts/1\n"
[WARNING] Debugger: Unexpected async-record: =cmd-param-
changed,param="inferior-tty",value="/dev/pts/1"
[Debugger] Log output: &"ptype TObject\n"
[Debugger] Log output: &"ptype Exception\n"
[Debugger] Log output: &"El s\303\255mbolo \302\253Exception\302\273
no est\303\241 en el contexto actual.\n"
TGDBMIDebugger.ProcessResult Error: ,msg="El s\303\255mbolo
\302\253Exception\302\273 no est\303\241 en el contexto actual."
[Debugger] Log output: &"ptype Shortstring\n"
[Debugger] Log output: &"ptype pointer\n"
[Debugger] Log output: &"ptype byte\n"
[Debugger] Log output: &"info file\n"
[Debugger] File type:
[Debugger] Entry point:
[WARNING] [Debugger.TargetInfo] Unknown FileType: , using GDB cpu
[Debugger] Log output: &"Funci\303\263n \302\253foo\302\273 no
definida.\n"
[Debugger] Console output: ~"[Depuraci\303\263n de hilo usando
libthread_db enabled]\n"
[Debugger] Console output: ~"Using host libthread_db library
\"/lib/x86_64-linux-gnu/libthread_db.so.1\".\n"
[Debugger] Console output: ~"[Nuevo Thread 0x7fffed2fa700 (LWP
5123)]\n"
[DBGTGT] *
[DBGTGT] running,thread-id="all"
[Debugger] Console output: ~"[Nuevo Thread 0x7fffecaf9700 (LWP
5124)]\n"
[DBGTGT] *
[DBGTGT] running,thread-id="all"
[WARNING] Debugger: Unexpected async-record: =breakpoint-modified,bkpt
={number="2",type="breakpoint",disp="keep",enabled="y",addr="0x0000000
000446e19",func="main",file="project1.lpr",fullname="/home/miguel/Prog
ramacion/LazarusTest1/project1.lpr",line="16",thread-
groups=["i1"],times="1",original-location="main"}
[WARNING] Debugger: Unexpected async-record: =breakpoint-modified,bkpt
={number="3",type="breakpoint",disp="keep",enabled="y",addr="0x0000000
000446e19",func="main",file="project1.lpr",fullname="/home/miguel/Prog
ramacion/LazarusTest1/project1.lpr",line="5",thread-
groups=["i1"],times="1",original-
location="/home/miguel/Programacion/LazarusTest1/project1.lpr:+0"}
[Debugger] Log output: &"info program\n"
[Debugger] Target PID: 5119
[Debugger] Log output: &"info address FPC_RAISEEXCEPTION\n"
[Debugger] Log output: &"info address FPC_BREAK_ERROR\n"
[Debugger] Log output: &"info address FPC_RUNERROR\n"
[Debugger] Log output: &"info line
\"/home/miguel/Programacion/LazarusTest1/unit1.pas\":35\n"
[WARNING] Debugger: Unexpected async-record: =breakpoint-modified,bkpt
={number="7",type="breakpoint",disp="keep",enabled="y",addr="0x0000000
000485316",func="EDIT1ENTER",file="unit1.pas",fullname="/home/miguel/P
rogramacion/LazarusTest1/unit1.pas",line="35",thread-
groups=["i1"],times="1",original-location="unit1.pas:35"}
[Debugger] Log output: &"kill\n"
[Debugger] Log output: &"info program\

----------------------------------------------------------------------
miguel-laz · 2013-04-30 19:26

The terminal displays the following message:

(lazarus:5110): LIBDBUSMENU-GLIB-WARNING **: Translation has an
invalid value 'I-->D' for default text direction.  Defaulting to left-
to-right.

----------------------------------------------------------------------
Martin_fr · 2013-04-30 19:43

Sorry, I dint see the attachment on the post earlier. reading that now

----------------------------------------------------------------------
Martin_fr · 2013-04-30 20:01

Quote from: miguel-laz on April 29, 2013, 09:07:15 pmJust installed
Lazarus 1.0.8 on Ubuntu 13.04 (64bit) and trying to debug a simple
program, the debugger stops at the breakpoint marked but when you stop
debugging (red square button) displays a debugger error window
indicating "The debugger entered into a failed state." In windows7
32bit works perfectly.

Ok, I went to through the log file (from the attachment. It looks like
everithing went correct.

When you stop (rud square) then the app is killed.
The IDE verifies that the app really is killed, by sending "info
program"

An English gdb will reply with a phrase containing "not being run" (as
the program is not being run), and the IDE checks for this text.
Your gdb responds  (as you actually pointed out correct)
Code: [Select]&"info program\n"
~"El programa que ha sido depurado no estaba corriendo.\n"

The IDE does not understand this, and assumes something went wrong.

Maybe this message was not translated in previous versions of gdb.
(you will notice, that some messages by your gdb are still English)

---
Another thing I spotted is, that the response to "info file" is not
English.

This does not cause an error right away. But it does mean, that the
IDE may not show you the type and msg of an "raise exception"

I know this problem also exists in older gdb, if they are translated.

----
and

""[Nuevo Thread 0x7fffecaf9700"

may mean that your thread window does not always update

----------------------------------------------------------------------
Martin_fr · 2013-04-30 20:02

Quote from: macmike on April 30, 2013, 07:55:22 pmI've been using
Lazarus since v9.3 on Ubuntu without problems until I upgraded to
Ubuntu 13.04 (32-bit) on the weekend and now I get this problem too

If yours is not translated, then there must be yet another issue. It
is possible the gdb team changed some of the english messages too.

----------------------------------------------------------------------
macmike · 2013-04-30 20:30

Quote from: Martin_fr on April 30, 2013, 08:23:35 pmQuote from:
macmike on April 30, 2013, 08:09:24 pmI'm using the English versions.
Where can I find the log file?
(sorry for the dumb question!)

As I wrote earlier in this thread:
http://wiki.lazarus.freepascal.org/GDB_Debugger_Tips#Log_info_for_debu
g_session

Sorry, I need to read more and type less 

I'm using Laz 1.0.8 (which is awesome cakes) FPC: 2.6.2 on i386-linux-
gtk 2

I set a breakpoint on some code and then when it gets invoked I'm
hovering over a variable, at that point a tooltip shows saying the
var=??? which is not surprising as it's a property with an accessor.
More importantly though the debugger appears to crash in that I can't
move in or over the next statement (F7,F8) or out of the method. In
fact pressing run F9 doesn't do anything either.

Pressing the stop button doesn't stop the debugging app and the
breakpoint no longer stops execution. Here's what I think is the
relevant bit of the log?

----------------------------------------------------------------------
Martin_fr · 2013-04-30 21:25

Quote from: macmike on April 30, 2013, 10:30:06 pm. More importantly
though the debugger appears to crash in that I can't move in or over
the next statement (F7,F8) or out of the method. In fact pressing run
F9 doesn't do anything either.

Pressing the stop button doesn't stop the debugging app and the
breakpoint no longer stops execution. Here's what I think is the
relevant bit of the log?

Indeed seems you found a bug in gdb. And gdb crashed:
Code: [Select] << TCmdLineDebugger.ReadLn
"&"/build/buildd/gdb-7.6~20130417/gdb/parse.c:588: internal-error:
mark_struct_expression: Assertion `parse_completion &&
expout_tag_completion_type == TYPE_CODE_UNDEF' failed.\nA problem
internal to GDB has been detected,\nfurther debugging may prove
unreliable.""
  [Debugger] Log output:
&"/build/buildd/gdb-7.6~20130417/gdb/parse.c:588: internal-error:
mark_struct_expression: Assertion `parse_completion &&
expout_tag_completion_type == TYPE_CODE_UNDEF' failed.\nA problem
internal to GDB has been detected,\nfurther debugging may prove
unreliable."
  << TCmdLineDebugger.ReadLn "&"\n""
  [Debugger] Log output: &"\n"

GDb may be able to recover and continue, after this crash (may be, or
may not be.)

The problem is that gdb does not print the prompt, that it is ready
for the next command. So the IDE waits indefinitely.


You may want to try, and change between stabs and dwarf (whatever you
currently use). Maybe that will fix the error.
If you do, ensure you do for *ALL* packages, that have debug info.


If it doesn't help, you can try and configure a timeout.
In the debugger page of the option dialog is a property grid. This has
an entry "TimeOutForEval" set it to 50. (And disable the
"WarnOnTimeOut" if you like.)

----------------------------------------------------------------------
miguel-laz · 2013-05-01 10:07

QuoteAn English gdb will reply with a phrase containing "not being
run" (as the program is not being run), and the IDE checks for this
text.
Your gdb responds  (as you actually pointed out correct)
Code: [Select]
&"info program\n"
~"El programa que ha sido depurado no estaba corriendo.\n"

The IDE does not understand this, and assumes something went wrong.
I can not understand that the answer is a text and not a numeric
identifier unique to the gdb reply message, do not understand!. Not a
criticism to the work done by the developers of Lazarus, is a thought.

I searched a parameter to force "gdb" to work internally with the
English language for message, but I have not found.
Could this be a workaround?

Should I report this bug somewhere else that Lazarus developers it
fixed in future revisions?
What I can do to fix the problem? Any solution?

----------------------------------------------------------------------
BigChimp · 2013-05-01 10:29

Quote from: miguel-laz on May 01, 2013, 12:07:05 pmI can not
understand that the answer is a text and not a numeric identifier
unique to the gdb reply message, do not understand!. Not a criticism
to the work done by the developers of Lazarus, is a thought.

Quote from: miguel-laz on May 01, 2013, 12:07:05 pmShould I report
this bug somewhere else that Lazarus developers it fixed in future
revisions?
What I can do to fix the problem? Any solution?

You could report it at the gdb bugtracker so they build in numeric
result codes, I'd say...

----------------------------------------------------------------------
Martin_fr · 2013-05-01 11:32

GDB provides a "machine readable" interface -mi
And Lazarus uses this.

But not all commands are available in this. So Lazarus still needs to
use some of the human commands.

Over the years GDB responses (English) from many platforms, and lots
of gdb versions have been collected. There are already many
differences there (last gdb 7.5 changed some spacing, and fooled the
IDE).

It will be near impossible to collect all that info, in dozens of
extra languages. And even if, it will be a
maintenance issue.

One could add translation files, which the IDE can load, and which the
user(s) can maintain. That would actually be a good idea. But I do not
know if or when I would have time for that.

----
For now: you can search
     debugger\gdbmidebugger.pp

It has all the text fragments. Search for "info program", "info file"

You will find several bits like this or similar
Code: [Select]  if CmdRes
  then CmdRes := ExecuteCommand('info program', R, [], 1500); //
Hardcoded timeout
  if (not CmdRes)
  or (Pos('not being run', R.Values) <= 0)
  then begin

You need to change the  (Pos('not being run', R.Values) <= 0)
to
Code: [Select](
   (Pos('not being run', R.Values) <= 0)
and
   (Pos('sido depurado no estaba corriendo', R.Values) <= 0)
)
or similar

Note: it is "and" here, because it is an error, only if both fail. In
other cases below, it may be "or"

Otherwise use your language only. But I advice to keep the english, in
case the text exists in more than one message, and is not always
translated.

----
The following is OK / IT is NOT yet translated
Code: [Select]    if ExecuteCommand('info program', [], R,
[cfCheckState])
    then begin
      s := GetPart(['child process ', 'child thread ', 'lwp '], [' ',
'.', ')'],

----
For info file
Code: [Select]    if ExecuteCommand('info file', R)
    then begin
      if rfNoMI in R.Flags
      then begin
        FileType := GetPart('file type ', '.', R.Values);
        EntryPoint := GetPart(['Entry point: '], [#10, #13, '\t'],
R.Values);
use "tipo de archivo "
and "Punto de entrada: "

Note spaces at end

Quote"S\303\255mbolos desde
\302\253/home/miguel/Programacion/LazarusTest1/project1\302\273.\n"
~"Local exec file:\n"
~"\t`/home/miguel/Programacion/LazarusTest1/project1', tipo de archivo
elf64-x86-64.\n"
~"\tPunto de entrada: 0x446cb0\n"


----
If you want threads to update *while* running (it is no problem to
skip that) also search for thread-created', 'thread-exited'

""[Nuevo Thread 0x7fffecaf9700"

----------------------------------------------------------------------
BigChimp · 2013-05-01 11:36

Quote from: Martin_fr on May 01, 2013, 01:32:32 pmGDB provides a
"machine readable" interface -mi
And Lazarus uses this.

But not all commands are available in this. So Lazarus still needs to
use some of the human commands.

Over the years GDB responses (English) from many platforms, and lots
of gdb versions have been collected. There are already many
differences there (last gdb 7.5 changed some spacing, and fooled the
IDE).

It will be near impossible to collect all that info, in dozens of
extra languages. And even if, it will be a
maintenance issue.

One could add translation files, which the IDE can load, and which the
user(s) can maintain. That would actually be a good idea. But I do not
know if or when I would have time for that.
Wouldn't it be much better to raise a bug with gdb that the -mi
interface needs to be extended?
----

----------------------------------------------------------------------
Martin_fr · 2013-05-01 11:45

Quote from: BigChimp on May 01, 2013, 01:36:03 pm
Wouldn't it be much better to raise a bug with gdb that the -mi
interface needs to be extended?


You can go and report them...

I actually need to check. Seems there are some new ones (maybe already
for a while) which the IDE does not yet use.


But the IDE also must maintain the old stuff, because on some
platforms only old GDB are available (Mac is is still at 6.3.50)

----------------------------------------------------------------------
BigChimp · 2013-05-01 12:53

Quote from: Martin_fr on May 01, 2013, 01:45:31 pmQuote from: BigChimp
on May 01, 2013, 01:36:03 pm
Wouldn't it be much better to raise a bug with gdb that the -mi
interface needs to be extended?


You can go and report them...
I'll leave that to Miguel as he has the actual problems...

Quote from: Martin_fr on May 01, 2013, 01:45:31 pmBut the IDE also
must maintain the old stuff, because on some platforms only old GDB
are available (Mac is is still at 6.3.50)
I understand.

----------------------------------------------------------------------
miguel-laz · 2013-05-01 16:20

QuoteYou could report it at the gdb bugtracker so they build in
numeric result codes, I'd say...
I do not feel able to report a problem I really do not understand
well, besides my English is basic, I use google-translator, I can
hardly express something so complex.

I think the problem should be addressed by the easy way, I think it'll
force "gdb" to that uses only English message strings.

I will continue looking for a solution this way.

----------------------------------------------------------------------
Martin_fr · 2013-05-01 16:58

You do not need to report them.  At the moment it would not benefit us
much.

Lot me know if the change to gdbmidebugger are helping you.

Or lets see if we can add info, how to get an english gdb. But that
would have to come from some one who knows the linux distro very good.

----------------------------------------------------------------------
miguel-laz · 2013-05-01 17:26

Finally I tried to send an email to gdb-patches@sourceware.org for
help, but it's not possible, "Invalid mime type" text / html "detected
in message text or attachment. Please send plain text messages only."

I do not understand anything.

----------------------------------------------------------------------
exdatis · 2013-05-01 17:32

Just set your mail client to send text messages?

----------------------------------------------------------------------
Martin_fr · 2013-05-01 17:41

Quote from: miguel-laz on May 01, 2013, 07:26:26 pmFinally I tried to
send an email to gdb-patches@sourceware.org for help, but it's not
possible, "Invalid mime type" text / html "detected in message text or
attachment. Please send plain text messages only."

I do not understand anything.

They have a web based bug tracker.

But without a list of thinks that we need, nothing will happen.

Even then, some one on their team would need to be interested to do
the work, or patches need to be supplied.

Trying to change gdb would be a very lengthy process. Unless we can
sent them patches.

----------------------------------------------------------------------
miguel-laz · 2013-05-01 17:51

Quote from: exdatis on May 01, 2013, 07:32:28 pmJust set your mail
client to send text messages?

I use Gmail, I've pasted the text as "plain text" and I marked the
message as "plain text" ....

It seems that this time it was sent, fingers crossed! and to wait for
the answer ...

----------------------------------------------------------------------
Martin_fr · 2013-05-01 18:06

gdb-patches is for patches not general question...

But besides that: I do not *know* how to change the gdb language. But
I *believe* it is an installation issue.  The language is probably
compiled in. So you need to re-install re-compile gdb. How exactly to
do that is probably something that depends on your linux distro (but I
do not know).

----------------------------------------------------------------------
Martin_fr · 2013-05-02 11:53

I saw the response on the gdb list.
QuoteIf Lazarus is depending on the language used by GDB to be
English,
it sounds like Lazarus should override the user's environment in
terms of i18n. Typically, the language is determined via the LANG
and/or LC_* environment variables, so it should be able to force
English by setting LANG/LC_ALL et al to Eg. "C" before starting GDB.

The quickest way to test this, is for you to set this before starting
lazarus.
If Lazarus gets the modified value, it should pass it on to gdb.
(afaik)

You can set Lazarus to force using your language in the IDE, even if
the environment is diff.

----------------------------------------------------------------------
miguel-laz · 2013-05-02 16:26

Quote from: Martin_fr on May 02, 2013, 01:53:08 pmI saw the response
on the gdb list.
QuoteIf Lazarus is depending on the language used by GDB to be
English,
it sounds like Lazarus should override the user's environment in
terms of i18n. Typically, the language is determined via the LANG
and/or LC_* environment variables, so it should be able to force
English by setting LANG/LC_ALL et al to Eg. "C" before starting GDB.

The quickest way to test this, is for you to set this before starting
lazarus.
If Lazarus gets the modified value, it should pass it on to gdb.
(afaik)

You can set Lazarus to force using your language in the IDE, even if
the environment is diff.

Right, this is the solution. I modified the environment variable
"LANG" to "export LANG = C" and debugging worked well, no errors.

Now we just need to automate the process, I thought of using "Project
Options-> Compilation-> (Execute Before: Conmand & Execute After:
Conmand)" but I'm not sure this works.

Do these commands correspond to terminal commands?Seems not!

How could I run "export LANG = C" before debugging and "export LANG =
es_ES.UTF-8" after debugging from Lazarus?

Any idea?

----------------------------------------------------------------------
Martin_fr · 2013-05-02 16:56

It will need to be set to TProcess.Environment.
But all the other env may have to be copied, or it may be lost to gdb.

Also I am not sure if C is always the right choice.

And this needs to be tested, how it affects other platforms....
Probably it will become, an option.

You can report it on mantis, so it will not be forgotten.

----------------------------------------------------------------------
miguel-laz · 2013-05-02 21:41

QuoteIt will need to be set to TProcess.Environment.
But all the other env may have to be copied, or it may be lost to gdb.

This solution is very interesting, I understand that the Lazarus 
creates process "gdb" and before starting the process "gdb" set
environment variables (copy of Lazarus + LANG modified process). When
finished debugging do nothing.

What Lazarus source file manages the creation and completion of the
process "gdb"?
Try to modify the source code to do some tests!

----------------------------------------------------------------------
macmike · 2013-05-02 21:56

Quote from: Martin_fr on April 30, 2013, 11:25:06 pmQuote from:
macmike on April 30, 2013, 10:30:06 pm. More importantly though the
debugger appears to crash in that I can't move in or over the next
statement (F7,F8) or out of the method. In fact pressing run F9
doesn't do anything either.

Pressing the stop button doesn't stop the debugging app and the
breakpoint no longer stops execution. Here's what I think is the
relevant bit of the log?

Indeed seems you found a bug in gdb. And gdb crashed:
Code: [Select] << TCmdLineDebugger.ReadLn
"&"/build/buildd/gdb-7.6~20130417/gdb/parse.c:588: internal-error:
mark_struct_expression: Assertion `parse_completion &&
expout_tag_completion_type == TYPE_CODE_UNDEF' failed.\nA problem
internal to GDB has been detected,\nfurther debugging may prove
unreliable.""
  [Debugger] Log output:
&"/build/buildd/gdb-7.6~20130417/gdb/parse.c:588: internal-error:
mark_struct_expression: Assertion `parse_completion &&
expout_tag_completion_type == TYPE_CODE_UNDEF' failed.\nA problem
internal to GDB has been detected,\nfurther debugging may prove
unreliable."
  << TCmdLineDebugger.ReadLn "&"\n""
  [Debugger] Log output: &"\n"

GDb may be able to recover and continue, after this crash (may be, or
may not be.)

The problem is that gdb does not print the prompt, that it is ready
for the next command. So the IDE waits indefinitely.


You may want to try, and change between stabs and dwarf (whatever you
currently use). Maybe that will fix the error.
If you do, ensure you do for *ALL* packages, that have debug info.


If it doesn't help, you can try and configure a timeout.
In the debugger page of the option dialog is a property grid. This has
an entry "TimeOutForEval" set it to 50. (And disable the
"WarnOnTimeOut" if you like.)

Thanks for this.

Changing from dwarf to stabs didn't appear to do anything so I went
back to dwarf. Now if I hover (after applying the suggested timeout
settings) and then try to step I get the Debugger Stopping errors
message and then:

Code: [Select]The debugger experienced and unknown condition:
Press "Ignore" to continue debugging. This may NOT be safe. Press
"Abort" to stop the debugger.
Exception: EAccessViolation with message "Access violation"
Context: TGDBMIDebuggerCommandEvaluate. State: Error

  $08060570
  $088F85F7  TGDBMIDEBUGGERCOMMANDEVALUATE__DOEXECUTE,  line 12357 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088F655B  TGDBMIDEBUGGERCOMMAND__EXECUTE,  line 10832 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E54EA  TGDBMIDEBUGGER__RUNQUEUE,  line 6557 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E3B24  TGDBMIDEBUGGER__UNLOCKCOMMANDPROCESSING,  line 6270 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E57D0  TGDBMIDEBUGGER__RUNQUEUE,  line 6614 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp

----------------------------------------------------------------------
Martin_fr · 2013-05-02 22:07

debugger\cmdlinedebugger.pp

the gdb stuff is in debugger\gdbmidebugger.pp  so it will be called
from there.

IIRC cmdlinedebugger.pp is also used for running projects if the
debugger is disabled (the unit for that is processdebugger , which is
not a real debugger)

----
Note that you will find environemntstuff, but that is nod for gdb. It
is stuff passed VIA gdb to the project (if set in Menu>Run>"run
parameters")


----
function TCmdLineDebugger.CreateDebugProcess(const AOptions: String):
Boolean;

This must also be fixed, because the current used CommandLine is
deprecated. exe-name and arguments must in future be separated.

For your own testing, you can set environment here.

But the final solution must ass a param to the function (maybe
TStrings) from gdbmidebugger

----------------------------------------------------------------------
Martin_fr · 2013-05-02 22:19

Quote from: macmike on May 02, 2013, 11:56:01 pm
Changing from dwarf to stabs didn't appear to do anything so I went
back to dwarf. Now if I hover (after applying the suggested timeout
settings) and then try to step I get the Debugger Stopping errors
message and then:

Code: [Select]The debugger experienced and unknown condition:
Press "Ignore" to continue debugging. This may NOT be safe. Press
"Abort" to stop the debugger.
Exception: EAccessViolation with message "Access violation"
Context: TGDBMIDebuggerCommandEvaluate. State: Error

  $08060570
  $088F85F7  TGDBMIDEBUGGERCOMMANDEVALUATE__DOEXECUTE,  line 12357 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088F655B  TGDBMIDEBUGGERCOMMAND__EXECUTE,  line 10832 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
 [/quote]

Ok, that is a bug in the Lazarus code.

I will need a new log from you please, so I can see which commands the
IDE sends to gdb before the IDE fails.

Also the log may tell me, if gdb is still reacting same, or if it has
taken damage to the debug session.

----------------------------------------------------------------------
macmike · 2013-05-02 23:00

Here you go:

Code: [Select]  OnIdle: UnLock
  TGDBMIDebugger.UnLockCommandProcessing: Execute RunQueue
    Executing (Recurse-Count=0) queued= 15 CmdPrior=0 CmdMinRunLvl=-1
: "TGDBMIDebuggerCommandEvaluate: n Thread=1, Frame=0" State=Pause
PauseWaitState=0
      >>Enter: TGDBType.ProcessExpression: state = gtpsInitial 
 Expression="n"
      <<Exit:  TGDBType.ProcessExpression: state = gtpsInitialSimple 
Result=False Kind=skSimple Attr= Typename="" InternTpName=""
TypeDeclaration=""
      >> TCmdLineDebugger.SendCmdLn "ptype n"
      << TCmdLineDebugger.ReadLn "&"ptype n\n""
      [Debugger] Log output: &"ptype n\n"
      << TCmdLineDebugger.ReadLn "&"No symbol \"n\" in current
context.\n""
      [Debugger] Log output: &"No symbol \"n\" in current context.\n"
      << TCmdLineDebugger.ReadLn "^error,msg="No symbol \"n\" in
current context.""
      TGDBMIDebugger.ProcessResult Error: ,msg="No symbol \"n\" in
current context."
      << TCmdLineDebugger.ReadLn "(gdb) "
      >>Enter: TGDBType.ProcessExpression: state = gtpsInitialSimple 
 Expression="n"
        ERROR: detected state loop in ProcessExpression
      <<Exit:  TGDBType.ProcessExpression: state = gtpsInitialSimple 
Result=True Kind=skSimple Attr= Typename="" InternTpName=""
TypeDeclaration=""
      >> TCmdLineDebugger.SendCmdLn "-data-evaluate-expression
parentfp"
      << TCmdLineDebugger.ReadLn "^error,msg="No symbol \"parentfp\"
in current context.""
      TGDBMIDebugger.ProcessResult Error: ,msg="No symbol \"parentfp\"
in current context."
      >> TCmdLineDebugger.SendCmdLn "-data-evaluate-expression 7"
      >> TCmdLineDebugger.SendCmdLn "-data-evaluate-expression 1"
      << TCmdLineDebugger.ReadLn "(gdb) "
      [WARNING] TGDBMIDebugger:  ExecuteCommand "-data-evaluate-
expression parentfp" failed.
      DebuggerState: Setting to Error, from Pause
        DebugDataMonitor: >>ENTER: TGDBMIThreads.DoStateChange  New-
State=Error
        DebugDataMonitor: <<EXIT: TGDBMIThreads.DoStateChange
        DebugDataMonitor: >>ENTER: TGDBMICallStack.DoStateChange  New-
State=Error
        DebugDataMonitor: <<EXIT: TGDBMICallStack.DoStateChange
        DebugDataMonitor: >>ENTER: TGDBMILocals.DoStateChange  New-
State=Error
        DebugDataMonitor: <<EXIT: TGDBMILocals.DoStateChange
        DebugDataMonitor: >>ENTER: TGDBMIWatches.DoStateChange  New-
State=Error
        DebugDataMonitor: <<EXIT: TGDBMIWatches.DoStateChange
        Canceling: "TGDBMIDebuggerCommandEvaluate: Result Thread=1,
Frame=0"
        Canceling: "TGDBMIDebuggerCommandEvaluate: y Thread=1,
Frame=0"
        Canceling: "TGDBMIDebuggerCommandEvaluate: x Thread=1,
Frame=0"
        Canceling: "TGDBMIDebuggerCommandEvaluate: n Thread=1,
Frame=0"
        >> TCmdLineDebugger.SendCmdLn "kill"
        DebugDataMonitor: >>ENTER: TSnapshotManager.DoStateChange 
New-State=Error
        DebugDataMonitor: <<EXIT: TSnapshotManager.DoStateChange
      DebuggerState: Finished Error
      DebugDataMonitor: TWatchValue.SetValidity: FThreadId=1 
FStackFrame=0 Expr=n AValidity=ddsError
      ERROR: Exception occured in
TGDBMIDebuggerCommandEvaluate.DoExecute EAccessViolation Msg="Access
violation" Addr=08060570 Dbg.State=Error
        $08060570
  $088F85F7  TGDBMIDEBUGGERCOMMANDEVALUATE__DOEXECUTE,  line 12357 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088F655B  TGDBMIDEBUGGERCOMMAND__EXECUTE,  line 10832 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E54EA  TGDBMIDEBUGGER__RUNQUEUE,  line 6557 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E3B24  TGDBMIDEBUGGER__UNLOCKCOMMANDPROCESSING,  line 6270 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E57D0  TGDBMIDEBUGGER__RUNQUEUE,  line 6614 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E5CC1  TGDBMIDEBUGGER__QUEUECOMMAND,  line 6696 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E68D6  TGDBMIDEBUGGER__GDBRUN,  line 6955 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $088E817E  TGDBMIDEBUGGER__REQUESTCOMMAND,  line 7370 of
/usr/share/lazarus/1.0.8/debugger/gdbmidebugger.pp
  $084E5374  TDEBUGGER__REQCMD,  line 6396 of
/usr/share/lazarus/1.0.8/debugger/debugger.pp
  $084E5A3B  TDEBUGGER__RUN,  line 6410 of
/usr/share/lazarus/1.0.8/debugger/debugger.pp
  $084D7414  TDEBUGMANAGER__RUNDEBUGGER,  line 2466 of
debugmanager.pas
  $084D1A24  TDEBUGMANAGER__ONRUNTIMER,  line 795 of debugmanager.pas
  $082C7545  TCUSTOMTIMER__DOONTIMER,  line 181 of customtimer.pas
  $082C74AF  TCUSTOMTIMER__TIMER,  line 156 of customtimer.pas
  $082BE1D4  GTKTIMERCB,  line 2852 of gtk2callback.inc
  $B7013107
      Canceling: "TGDBMIDebuggerCommandEvaluate: n Thread=1, Frame=0"
      Canceling: "TGDBMIDebuggerCommandEvaluate: n Thread=1, Frame=0"
      DebuggerState: Setting to None, from Error
        DebugDataMonitor: >>ENTER: TGDBMIThreads.DoStateChange  New-
State=None
        DebugDataMonitor: <<EXIT: TGDBMIThreads.DoStateChange
        DebugDataMonitor: >>ENTER: TGDBMICallStack.DoStateChange  New-
State=None
        DebugDataMonitor: <<EXIT: TGDBMICallStack.DoStateChange
        DebugDataMonitor: >>ENTER: TGDBMILocals.DoStateChange  New-
State=None
        DebugDataMonitor: <<EXIT: TGDBMILocals.DoStateChange
        DebugDataMonitor: >>ENTER: TGDBMIWatches.DoStateChange  New-
State=None
        DebugDataMonitor: <<EXIT: TGDBMIWatches.DoStateChange
        DebugDataMonitor: >>ENTER: TSnapshotManager.DoStateChange 
New-State=None
        DebugDataMonitor: <<EXIT: TSnapshotManager.DoStateChange
      DebuggerState: Finished None
    Exec done
    Leaving Queue with count: 0 Recurse-Count=0 State=No

----------------------------------------------------------------------
Martin_fr · 2013-05-03 00:05

Unfortunately the log does not go back to the point where the gdb
crash happened. So it is hard to interpret it.

It may however be enough to figure out what crashes the Lazarus exe.

However that will probably not help to much. It does look that gdb is
no longer fully responsive at this stage.

The IDE detects that via the timeouts. Unfortunately the timeout does
not recover. As a result the IDE does the last thing it can do: Go to
error state, and try to at least shut down the debug session. However
going to error state at this particular location leads to a crash.
That means fixing the crash will only allow the IDE to inform you that
the debug session was lost.

BTW: longer timeouts will (most likely) not help. the 50 milliseconds
you specified only indicate how soon the IDE starts detection.
Detection will wait about 3 full seconds minimum.

I can see you are hovering (or add a watch) over "n". But there is no
such variable.
"parentfp" is an internal symbol. It is used to detect if or if not
you are in a nested procedure (you probably were not). On that gdb
failed, which is very unusual.

------------------

I have seen a similar issue before.  If gdb fails on looking for a
symbol that does not exist, then it means that gdb must scan the
*entire* list of *all* symbols in the exe.
And if just one symbol in that list is not good to gdb, then good bye.

** 1)
Do you use any extra packages? bgra, lazreport, tachart, 3rd party?

Each package has its own settings for dwarf versus stabs.
Unfortunately the IDE is not smart enough to detect if a project mixes
different settings.

So your final exe can end up with both. Technically that is possible.
But gdb may dislike it (some gdb seem fine with that, but not all).

So please check each used package.

** 2)
This is only to be checked in the project. on the linking page of the
options.

Make sure you do NOT smart link (units can be "smart linkable", but
the linker must NOT smart link them)

Some linkers, when smart linking include stuff for code that is not in
the exe, and that can also cause gdb to get into trouble.

--------------------------------------
If neither of that helps: Did the previous/older gdb do better?

----------------------------------------------------------------------
miguel-laz · 2013-05-03 11:43

Well, this is the first approximation of source code modification of
the method TCmdLineDebugger.CreateDebugProcess () but there are a
number of issues that I have not yet very clear ...

1. SysUtils.GetEnvironment? .... () Refers to the main application
process?I understand that it is!

2. The format of each entry in TProcess.Environment think is right, I
need to see the value of TStrings doing a debug .

Code: [Select]function TCmdLineDebugger.CreateDebugProcess(const
AOptions: String): Boolean;
var
  envCount: integer;
  cont:integer;
  envValue:string;
  checkLang:boolean;
  setLangValid:boolean;
begin
  if FDbgProcess = nil
  then begin
    setLangValid:=False;
    FDbgProcess := TProcess.Create(nil);

    // Get all environment variables ¿current application process?
    // and add to FDbgProcess.Environment
    envCount := SysUtils.GetEnvironmentVariableCount;
    if envCount > 0 then
    begin
      for cont:=1 to envCount do
      begin
           envValue:=SysUtils.GetEnvironmentString(cont);
           checkLang:=StrUtils.AnsiStartsText('LANG=en', envValue);
           if checkLang = True then
           begin
              envValue:='LANG=C';
              setLangValid:=True;
           end;
           FDbgProcess.Environment.Add(envValue);
      end;
    end;

    if setLangValid = False then
       FDbgProcess.Environment.Add('LANG=C');
    // end LANG case
    "
    "


How I can debug Lazarus himself?

----------------------------------------------------------------------
Martin_fr · 2013-05-03 12:09

Quote from: miguel-laz on May 03, 2013, 01:43:41 pm
How I can debug Lazarus himself?

Open: ide/lazarus.lpi

----------------------------------------------------------------------
miguel-laz · 2013-05-03 16:03

Quote from: Martin_fr on May 03, 2013, 02:09:10 pmQuote from: miguel-
laz on May 03, 2013, 01:43:41 pm
How I can debug Lazarus himself?

Open: ide/lazarus.lpi

I tried to debug Lazarus "ide/lazarus.lpi" from a portable version
(usb pen drive) Lazarus and does not work, now on a version installed
everything worked properly ... thanks

It is really hard to debug Lazarus itself! In Ubuntu, I had to create
a local installation from the sources that the original location
"usr/share/lazarus" I have permission problems.

When you start debugging an error "in dir EReadError 8A6463" debugging
options step-to-step all disabled on the primary IDE, the debugger
stops at the breakpoint selected but does not show the values ​​of the
variables.

Does the IDE in debug runs before leaving
TCmdLineDebugger.CreateDebugProcess ()? Can more than one process?

Developers Lazarus! Like what are you doing?

----------------------------------------------------------------------
miguel-laz · 2013-05-03 18:19

It seems the problem has been resolved. 

The debugger error window no longer appears at the end of debugging
with the stop button (red square), the method
TCmdLineDebugger.CreateDebugProcess () looks like this:

Code: [Select]function TCmdLineDebugger.CreateDebugProcess(const
AOptions: String): Boolean;
var
  envCount: integer;
  cont:integer;
  envValue:string;
  checkLang:boolean;
  setLangValid:boolean;
begin
  if FDbgProcess = nil
  then begin
    setLangValid:=False;
    FDbgProcess := TProcess.Create(nil);
    FDbgProcess.CommandLine := UTF8ToSys(ExternalDebugger + ' ' +
AOptions);
    // TODO: under win9x and winMe should be created with console,
    // otherwise no break can be sent.
    FDbgProcess.Options:= [poUsePipes, poNoConsole, poStdErrToOutPut,
poNewProcessGroup];
    FDbgProcess.ShowWindow := swoNone;
    AssignUTF8ListToAnsi(DebuggerEnvironment,FDbgProcess.Environment);
  end;
  if not FDbgProcess.Running
  then begin
    // Get all environment variables ¿current application process?
    // and add to FDbgProcess.Environment
    envCount := SysUtils.GetEnvironmentVariableCount;
    if envCount > 0 then
    begin
      for cont:=1 to envCount do
      begin
           envValue:=SysUtils.GetEnvironmentString(cont);
           checkLang:=StrUtils.AnsiStartsText('LANG=', envValue);
        if checkLang = True AND
           StrUtils.AnsiStartsText('LANG=en', envValue) = False then
           begin
              envValue:='LANG=C';
              setLangValid:=True;
           end;
           FDbgProcess.Environment.Add(envValue);
      end;
    end;

    if setLangValid = False then
       FDbgProcess.Environment.Add('LANG=C');
    // end LANG case

    FDbgProcess.Execute;
    DebugLn('[TCmdLineDebugger] Debug PID: ',
IntToStr(FDbgProcess.Handle));
  end;
  Result := FDbgProcess.Running;
end;

this is the new log file "laz.log" (attached), everything seems fine.

In short I will create a new entry error report "bugtracker".

----------------------------------------------------------------------
Martin_fr · 2013-05-03 18:38

Good, but as I explained: To make it into the official lazarus, it
must be optional, and it must be limited to gdb based debuggers.

Anyway, glad it works for you now.

----------------------------------------------------------------------
miguel-laz · 2013-05-06 19:11

Just wanted to mention that I received an email from "Tom Tromey" of
"Redhat" says:

Code: [Select]I just wanted to mention that it would be even better to
change Lazarus
to use MI.  This is what MI was made for.
To solve this problem Do I need to create an entry in Bugtracker?
Is it enough grounds here?

----------------------------------------------------------------------
Martin_fr · 2013-05-06 20:08

We use MI.

But not all commands work in MI . And even those that do, we still
need to support old ones.
Though to be fair, some more commands could get MI fallback. But it's
a lot of work...

You can create a bug report, for the setting of LANG=C

----------------------------------------------------------------------
Martin_fr · 2013-05-09 22:47

@macmike

The problem you have may be the same as referenced here:
http://wiki.lazarus.freepascal.org/GDB_Debugger_Tips#Issues_with_GDB_7
.5.9_or_7.6

----------------------------------------------------------------------
macmike · 2013-05-10 00:34

Thanks for the link - sorry I've not come back to this I've just
bought a mac so I've been a bit focussed on getting that set up with
Lazarus (which was easy and the results are brilliant).

I've had a bit of luck stabilising the debugger by fiddling with the
options... for some reason I had -gh (heaptrc for memleaks) and -pg
(gprof) on. I've noticed for a while that the memleak report was
always showing 0 even if I tried to deliberately cause a leak so I was
fiddling around with these settings. Long story short, memleak
reporting is now working and the debugger can step on but it still
crashes after hovering over (almost) any value.

I tried removing all of my watches and hiding the local vars window
but it didn't help. My code's heavily instrumented so I'm kind of
getting away with it at the moment. Also, I'm building on Linux, Win
and Mac so I've got a choice of platforms to play with.

I had a theory it could be do to with my code somehow so I created the
following test app which causes the crash after hovering over
OpenDialog1.FileName (hovering over "blah" is ok):

Code: [Select]procedure TForm1.Button1Click(Sender : TObject);
var blah : string;
begin
  blah := 'hello';
  OpenDialog1.Execute;
  if OpenDialog1.FileName = blah then
    Exit;
end;      

Full log for the session attached.

----------------------------------------------------------------------
Martin_fr · 2013-05-10 00:49

The logfile starts with

Quote"GNU gdb (GDB) 7.5.91.20130417-cvs-ubuntu\n""

And due to the bug in GDB this version of GDB can not be used.

You will need to install an older version of GDB.
Or if you build it yourself, you can try the patch from the gdb bug
report.

----------------------------------------------------------------------
macmike · 2013-05-10 11:01

Edit: Spoke to soon.... Reinstalling gdb made it a bit more stable but
didn't fix the problems entirely. Off I go to downgrade gdb.

----------------------------------------------------------------------
macmike · 2013-05-10 12:56

I've downgraded gdb to 7.5-ubuntu and now debugging seems stable
again. Thanks for your help